Correlation analysis of solar power and electric demand

Solar power is one of the fastest growing renewable energy source around the world. Since the peak solar power produced is maximum during the mid-day and also the load demand is at its peak around this time, solar power is a suitable alternative source of electricity with less cost and pollution. This paper investigates the relationship between solar power and electric demand. Mathematical models using Artificial Neural Networks (ANN), Fuzzy Inference System (FIS) and Adaptive Neuro Fuzzy Inference System (ANFIS) techniques developed based on the weather data and historical load data were used to forecast electricity demand in Singapore. Both day-ahead average load and hourly load forecast for the day was done using different models and different test sets. Mean Absolute Percentage Error (MAPE) for each model was calculated to evaluate the performance and compare with other models. Solar power generation in Singapore was estimated using actual weather data, and the correlation studies between the electric demand and solar power generation were carried out, proving it to be a potential substitute for power generarion duriny day time.